A note on some observations of dye in coastal waters

Columbia University in the City of New York

LAMONT GEOLOGICAL OBSERVATORY
PALISADES, NEW YORK

A NOTE ON SOME OBSERVATIONS
OF DYE IN COASTAL WATERS

Report prepared by: R. Gerard

and

B. Katz

Technical Report No, CU-3-63 to the Atomic Energy Commission

Contract AT(30~1)2663

July, 1963

A NOTE ON SOME OBSERVATIONS
OF DYE IN COASTAL WATERS

Report prepared by: R. Gerard

and

B. Katz

Technical Report No, CU-3-63 to the Atomic Energy Commission

Contract AT(30-1) 2663

July, 1963

This publication is for technical information only and does not
represent recommendations or conclusions of the sponsoring
agencies, Reproduction of this document in whole or in part is
permitted for any purpose of the U.S. Government,

In citing this manuscript in a bibliography, the reference should
state that it is unpublished.

-1-

In his recent text (1962) von Arx has written, (p. 114) '‘The Ekman spiral
has been observed in the atmosphere and has been produced in the laboratory in
rotating tanks and ocean models, but its occurrence in the wind -influenced layer
of the ocean has not been demonstrated beyond the tendency for sea ice and some
currents to move at 6ome angle to the right of the wind*" In a number of observa*
tions using dye tracers we have noticed certain features which seem to be explained
best by the "Ekman Effect", though the circulation pattern in the area of these
experiments is so intricate that it is doubtful whether an Ekman spiral in the ideal
sense could be expected*

Since the summer of 1961 we have performed a number of experiments usin
Rhodamine dye on the sea surface as a tracer to aid in the study of turbulent
diffusion and local circulation* The methods used are those described by Pritchard
and Carpenter (I960)* We often supplement our measurements with aerial photo¬
graphs, mainly to assist in determining the actual shape of the dye patch, which is
often fairly complex*

Most of our observations have been made in the area of the New York Eight*
Here the circulation in summer and fall is dominated by weak counter-clockwise
eddies close to the New Jersey shore, which carry southward the low salinity
effluent of the Hudson -Raritan estuaries* Outside this coastal band, about ten
miles off shore in more saline water, there is a northward moving current* Super¬
imposed on this pattern are weak tidal currents whose main vectors lie on a north¬
west - southeast axis, the excursion of the ebb to the southeast being of greatest
magnitude* This circulation pattern is rapidly broken up in the presence of storms*
The wave and swell pattern may be very complex due to reflection and refraction
at the shores, and refraction over the Hudson submarine Canyon*

Our discussion concerns certain characteristics of surface dye patches in

i. '. V T ' (- 1

V- j 4

; • . , . .* ' • i it : '

'

-

• .

i t . • - - - /

c ?. r!

rn • ;tu - r. • r. • ■ « - •

;

•: ’

. * *

Dbon . vwH sw 9100*1* ir\ ; . b

« •

(I jl * ■>

. - ? • \ : > o*f? : . : rn r:>L t>. j \d Sc <

* f, t < • ■ ; ; " . ‘ •. N-' i : I .

a. . ■ ..?• .

. . . •' . , .v • • * • • • ■ . • ! '• ! ' ' . -..i. . • . • £♦ =:d"

_•! ‘ v.

, . • r , t • . .

- V*

v • f • • .... . /• . . » 1 . f-M' • } i>j r . 't *, j un:

d : 1 ! . ' • ’

4 • i ■

1 • . . .

: ; r-K . ! tdrr.-.. : -.U, . - " J

\ • • • : ,3101 fdlia . ' 3r *- :

, • •' . , r- : ? ’ • i i -.-.. • 4-. 9tU

• W , i. .• . .• : • * ‘ ... ..'••V sn'fsr. • . r:i s‘i -J ■ Vv • tlm

: j • • • ' . .. £.• , , la.. • • • ; .

■

» • .

• . . . • • ‘ . *.*Ci . .

' • • ' L:

-2-

this area, particularly during the early stages of their spreading* Figures 1, 3
and 5 are aerial photographs of dye patches on different dates; all exhibit features
which appear to be common during the first few hours of surface dye introductions
in the presence of steady, moderate winds and in the absence of strong currents and
high seas* These characteristics can be listed as follows:

1* The patch will initially take on a striated pattern which is
aligned with the swell waves.

2, The trailing (more tenuous) portion of the patch that most
strongly exhibits the striated pattern lies at a level below
the surface, while the concentrated leading portion lies

at the surface. This can be clearly seen in Figure 1, where
the research vessel (whose draft is nine feet) has stirred
dye up to the surface when passing through the tail (center
of photograph). On the upper right in this photograph,
the ship1 s wake reveals that it has crossed the head of the
dye and stirred clear water up to the surface.

3. Where moderate winds prevail, the dye patch takes a comet¬
like form with a curved tail. In all but one case the curva¬
ture was counter-clockwise, i. e. as though the leading
portion (the high concentration head) was turning to the left.

The curvature has been observed in seven out of ten experiments, six of
which were conducted in the New York Bight, two south of Jamaica in the

Caribbean, and two on the Eahama Banks. Of the three experiments which did not

produce a curved pattern, two were in the very shallow water on the Bahama Banl
the other, south of Jamaica, consisted of a long line of dye introduced fifteen feet
below the surface and running perpendicular to the direction of the wind, which

• •••-* -r;-v . >

: . -■ ’ .. .•••• • •

>r’ ’ • ■■ ' 1. •; . .. „ i > .

; •

■

; ’ . V ...

'

' 3* £ . s %'/: l . • ,

- ! ' .

*

f

i . • c 5. •; !

- ■' :i '■

'■ '

’■ • • ' ' • . • ' ' ' ... -

• . bt e to ... 5fiJ a ...

■ )

o,'.: aria

' '

• ' . ..

.

• • ' • " ' •••• ‘ • . • > . ;

'

■ * ' • ■ .■ : .. . . .
* • r •

■ ■

‘ • ' •• ■ • ' • . ■ .

■ • • . ;

‘ a ■ '■

.

■ - ■

-3 —

was mild (about 5-10 knots), until the very last stages when the dye had already
become highly attenuated.

Since we have observed that in surface and near-surface experiments the
dye responds quite readily to changes in the wind, we attempted at first to explain
the curvature on this basis. However, in four cases where no significant change in
wind direction was observed, the dye patch assumed a counter-clockwise curvature.
Thus changes in wind direction are not of themselves adequate explanation for the
observed curvature.

If local and/or transient gyrals are the cause of the curvature, we should
reasonably expect that they would have random directions of rotation, and hence
that we should see clockwise and counter-clockwise curvatures with about equal
frequency. In fact, we have observed clockwise curvature only once in a dye
introduction at fifteen feet depth and there is evidence for attributing this case to
the effects of changing wind direction, (The wind had shifted from SSE to NW),

The persistence of this counter-clockwise curvature in the presence of moderate
winds which remain steady in direction makes it unlikely that it can be attributed
to purely local or transient phenomena.

The shear between horizontal water layers evidencedin photographs, such

as Figure 1, suggests that the pattern we see could be the result of rotary tidal

currents or of wind-driven drift currents. Inf airly shallow waters cum sole rotary

tidal currents could produce patterns similar to the ones we have observed, but in

inshore waters reflection from the coastline and the effects of bottom irregulari-
c an

ties combine with tidal forces to give alternating or contra solem rotary tidal
currents. Added to this the effluent of large rivers can produce very distorte d
tidal current ellipses. Such is the case in the New York Bight, the extreme com¬
plexity of which is indicated in the tidal current diagram Fig, Z , It seems unlikely

Digitized by the Internet Archive
in 2020 with funding from
Columbia University Libraries

https://archive.org/details/noteonsomeobservOOgera

-4-

that such a current pattern could account for the observed shapes of our dye patches.
Furthermore, if the dye pattern is due mainly to tidal currents, in an experiment
conducted over two or more tidal cycles, a periodic change of shape is expectable.
No such periodicity has been observed. There remains to be considered pure
wind effects. The ready response of the dye to wind changes, the general resem¬
blance of the patterns to the upper portions of an Ekman spiral, and the fact that
fewer difficulties arise from attributing the curvature to the combination of wind
stress and Coriolis force than any of the other possibilities considered make us
favor this explanation.

Projected on a horizontal plane, a wind-driven current will show a
decrease in velocity and change of direction at regular intervals of depth down to
a depth D, the upper frictional depth. The classical relationship is:

D = 7.61 w si n<j> meters, where w = wind velocity and (p = latitude, but most
investigators take it to be about 100 meters in mid-latitudes and for moderate
winds (Ekman, 1928; Proudman, 1953). Figures 3 and 5 are photographs of
experiments conducted in different areas within the New York Bight. The photo¬
graphs were taken about one and one -half and two hours respectively after the dye
was introduced.

At the time of these experiments the seasonal thermocline had disappeared,
and temperatures were nearly uniform to the bottom. The depth in the areas of
these experiments was considerably less than 100 meters, and the effects of the
bottom friction have been taken into consideration in the manner of Defant (1961)
in determining the configuration of the expected distribution. To obtain the
expected configuration, we have assumed an established drift current having

velocity at the surface given by Proudman (1952:

Tpk r*

us =

u

y/Z D Y*

-*• ’>y * ; 1 • • - . • • , • .f.t

: - */ J , ' :

■

^ ‘ • ' ... . ‘ w. - 1 Vj

• .. . u:
j VI Jr: n

where:

-5-

= I I _ , the half pendulum day

w sin 0

= 2.5 x 10~3

= density of air, taken as 1.25 x 10"^ gm/cm^

= density of sea water, taken as 1.03 gm/cm^
a thickness of the upper frictional layer, taken as 100 meters
= the wind speed.

Using this formula, we have plotted the position of a "disk” of dye on the surface
two hours after dye injection. From a graphical description by Defant giving the
vertical structure in drift current for water depths less than D, we have been able
to plot the relative positions of several other '’disks" at discrete levels down to
forty feet, the limit of visibility. The size of these "disks" was estimated on the
basis of a "most probable diffusion velocity of 1 cm/ sec, which is approximately
the value yielded by our experiments. Distortion due to the fact that the dye does
not instantly reach a depth of forty feet is believed to be minimal, since vertical
diffusion was quite rapid due to the homogeneity of the water. In plotting the
observed configuration, the drift due to tidal motions has been estimated from tide
tables and tidal current charts, and this has been subtracted from the actual dis¬
placement of the dye from the pcsition at which it was dumped. Shear between
horizontal layers due to tidal movements, the river effluent, and the intricate
pattern of the coastline could also be expected to contribute their effects. However;
since we have insufficient data, these influences have not been taken into account.

The experiment on October 17 (Figures 3 and 4) was made about eight
miles east of Sandy Hook in water of 27 meters depth. The wind was fairly steady
for several hours at about 16 knots from the west. The observed displacenment
of the dye patch was to the left, rather than to the right of the wind. Since only tidal

k

r*

r

D

Ua

■

'

.

.

• :

Miv .itt

Oct. 17, IS 61

At Ambrose Lightship
2 hours after dump
27. meters deep

DUMP

O

0.2 mi

-6-

drift was taken into account in the figure, these seemingly contradictory results couli
readily be accounted for by a net north current of 0,2 knots. The counter-clockwise
curvature is still present, but greatly distorted by other forces.

On October 31 at a location about nineteen miles southeast of Sandy Hook,
two experiments were performed in water of 61 meters depth. In the first, a
barrel of dye was introduced at 0300 hours, when the wind was 7 knots from the
we st- southwest. At 0900 hours the wind remained unchanged, and at 0800 hours
the dye had roughly assumed the shape of a broad ellipse with many finger-like
projections perpendicular to the long axis on both sides. This shape was main¬
tained until about 10:30, at which time the second experiment was begun. The
second dye patch immediately began to assume a comet-like shape with tenuous
tails and counter-clockwise curvature, A wind reading taken at this time showed
that the velocity had increased to 12 knots, still from the west. Figures 5 and 6
show the pattern of the second experiment about two hours after the introduction.

At this time the plane returned to the site of the first dye patch to find that it, too,

■s «»• 1

was beginning to assume a shape with an unmistakable counter-clockwise curvature.

A number of photographs were taken showing this, but the dye had become so
attenuated that good reproductions of the photographs cannot be made. The photo¬
graphs show that, in both cases, the orientation of the curved dye patch with
respect to the wind was the same and about what would be expected for a pure drift
current.

In later stages of dye patches we have observed that, when the direction
of the wind is steady, the curvature usually tends to disappear. We believe that
this is due to the fact that as the dye becomes more attenuated, the depth to which
it can be seen decreases. Thus, although the overall dimensions of the pattern
increase with time, we observe an increasingly smaller portion of an increasingly

.

*

t ••

FIGUr

Inlet

Fig. 8

Oct. 31,196!

10 mi. east of

Shark
d u m c

i

ep

\

U.i mi.

K-

-7-

larger spiral.

Probably the best evidence in support of the notion that the observed
curvature is the result of the Ekman effect is its persistence in a wide variety of
circumstances. It would, therefore, be interesting to see the results of a similar
experiment conducted in the southern hemisphere, where the curvature should be
clockwise.

It is also of interest to consider a dye experiment designed to test the
wind current. For quantitative work, an area of simple circulation and greater
depth would have to be chosen. A useful method might be the simultaneous ejection
of filaments or pulses of dyd at regular depth intervals in a clear deep-water area.

The resulting pattern could be photographed from the air under various wind condi¬
tions, Several such experiments are underway at the present time.

■

.

OjO

* CW - Clockwise
** CCW - Counter-clockwise

!

14:06 135

(cont'd) 4

»

IV

f

I

i

&

i

)

!

<i

References

Ekman, V. W. ; A Survey of Some Theoretical Investigations on Ocean-Currents,
J. du Conseil, Vol. III, No. 3, December, 1928.

Pritchard, D. W. and J. H. Carpenter, Measurements of Turbulent Diffusion in
Estuaries and Inshore Waters: Bull. Int. Assoc. Sci. Hydrol.,

No. 20, 37, I960.

Proudman, J.; Dynamical Oceanography, J. Wiley & Sons, New York, 1953.

Defant, A,; Physical Oceanography, Pergamon Press, New York, Vol. I, 1961.

Leaf, W. B.; ’’Tracing Water Movement," Undersea Technology, pp. 24-26,
March, 1963.

von Arx, W. S.; An Introduction to Physical Oceanography, Addison-Wesley,

Reading, Mass., 1962.

ACKNOWLEDGMENT

The work reported in this paper was supported by the Atomic
Energy Commission of the U.S. Government under Contract AT(30-1)2663.

i

-

►

, T

’

1

* .

1

COLUMBIA LIBRARIES OFFSITE